In a global crisis, like this pandemic, it's not a difficult task to take a moment, look around and observe other people's reactions. Most of us strive to get information about the COVID-19, countless times per day, about what is happening, how it will affect our future or when we will get back to our regular lives. We realised that the most crucial thing in a situation like this is how you deliver the message. We all agree that news travels extremely fast, and we need valid information in order to make decisions and plans. We believe in the simple, not the complex. We believe that we need to own and control the way we get informed.
That's why we came up with an initiative to empower every person to have a personalised news hub, where they can come to find and discover unbiased COVID-19 updates without being confused by the massive amount of information or getting stressed by the overwhelming content.
We introduce "Covid Whisperer".
The solution we bring to the table
Covid Whisperer is a news platform that provides COVID-19 related content based on fake news detection and sentiment analysis in combination with the user's personal preferences. The envisioned solution can be considered a hub, where news articles from a large number of sources are aggregated and displayed in a comprehensive and meaningful way for each user, with the assistance of intelligent algorithms.
More specifically, this set of algorithms are responsible for providing:
- up-to-date and reliable content related to the COVID-19 pandemic from various sources.
- intelligent techniques for the detection of fake news in the displayed content.
- indicators of emotional vocabulary in the displayed content.
The core algorithmic functionality will be provided via a number of independent back-end components, which will communicate with the external systems via a dedicated API. The incoming load of news articles will be stored in Elasticsearch engine, which also provides search and analytics capabilities in near real-time. As far as the front-end is concerned, the team's aspiration is to build Covid Whisperer both as a web and a mobile application using state-of-the-art technologies.
What we have done during the weekend
During these last couple of days, we have started from scratch and, through lots of brainstorming and alternating standpoints, our team ended up with:
- A well-defined idea, verified as a need by a targeted user research and analysis of the market competition.
- A set of novel features that support the implementation of our idea, backed by research in different areas (covid-19 crisis, fake news, sentiment analysis, personalised information, news platforms).
- A proposed architecture for the envisioned solution with details about each participating component, and alternatives.
- An indicative demo, showcasing the most significant features of the application.
- A solid business model with details about the feasibility, requirements and economic plan of our solution, supported by a complementary SWOT analysis.
- An early implementation of the various components of our application, i.e. data storage, back-end and front-end, which will serve as the starting point for an end-to-end implementation. The code is available in the project's Github repository.
The solution’s impact to the crisis
Our solution's impact to the crisis is expected to be threefold:
- Provide users with an environment of trust, where they can find reliable information about COVID-19 in an easy and fast way, according to their own preferences.
- Improve the user experience and quality of service by providing indicators that are related with the articles' contents, such as fake news and emotional vocabulary indicators.
- Provide statistics related to the trending news topics, as well as analytics related to the amount of fake news and sentimental content over time and per topic.
We could argue that all of the above essentially converge to the ultimate goal of bringing everyone closer to the truth about the events taking place in their communities and provide them with the freedom to consume the content of their choice.
The necessities in order to continue the project
Planning ahead, our team has identified the following key-necessities moving forward:
- Implementation of a MVP, based on the current feature set, and distribution to a group of users for testing purposes.
- Gathering of user feedback and repetition of user research for bug fixes and addition of extra features.
- Collaboration with UX designers/experts for the improvement of the user experience and quality of the final product.
- Collaboration with social scientists in the establishment of an optimum model for sentiment analysis.
The value of our solution after the crisis
The need for unbiased, reliable sources of information goes beyond the times of crisis we are experiencing right now. Click-baits, misleading or biased content and propaganda are only a few of the issues that have resulted in the mistrust and depreciation of information media. We strongly believe that it is essential to do our best to bridge that gap.